In 24 consecutive patients with documented ventricular tachycardia (VT) (22 patients) or fibrillation (VF) (2 patients), results of 2 programmed ventricular stimulation protocols to initiate VT/VF were prospectively studied. Seventeen patients had VT/VF after a healed myocardial infarction (MI) and 7 patients had idiopathic VT. In both protocols (designated 1 and 2), the right ventricular (RV) apex was paced at 100 beats/min, using a maximum of 2 ventricular premature complexes (VPCs) given at twice diastolic threshold. This protocol had a sensitivity of 25%. In protocol 1, the pacing site was changed to the RV outflow tract and the previous steps were repeated; in protocol 2, the pacing rate was increased to 120 and 140 beats/min at the RV apex, also using a maximum of 2 VPCs. The next step in protocol 1 consisted of increase of current strength to 20 mA and repeating previous steps at the RV apex and RV outflow tract, with a maximum of 2 VPCs; in the next step in protocol 2, three VPCs were used during sinus rhythm and pacing was performed at rates of 100, 120 and 140 beats/min. In protocol 1, therefore, only stimulation site and current strength were changed, while in protocol 2 only pacing rate and number of VPCs were modified. Protocol 1 had a sensitivity of 54% and protocol 2 a sensitivity of 83%. The sensitivity of protocol 2 was statistically higher than that of protocol 1 (p less than 0.05). In the group of patients with VT after MI, the sensitivity was 66% for protocol 1 and 93% for protocol 2.(ABSTRACT TRUNCATED AT 250 WORDS)